Advanced

Model-based analysis and optimization of a full-scale industrial high-rate anaerobic bioreactor

Feldman, Hannah; Flores-Alsina, Xavier; Kjellberg, Kasper; Jeppsson, Ulf LU ; Batstone, Damien J. LU and Gernaey, Krist V. LU (2018) In Biotechnology and Bioengineering p.2726-2739
Abstract

The objective of this paper is to present the model-based optimization results of an anaerobic granular sludge internal circulation reactor. The International Water Association Anaerobic Digestion Model No. 1 extended with phosphorus (P), sulfur (S), and ethanol is used to describe the main biological and physico-chemical processes. The high-rate conditions within the reactor are simulated using a flow + reactor model comprised of a series of continuous stirred tank reactors followed by an ideal total suspended solids separation unit. Following parameter estimation by least squares on the measured data, the model had a relative mean error of 13 and 15% for data set #1 and data set #2, respectively. Response surfaces show that the... (More)

The objective of this paper is to present the model-based optimization results of an anaerobic granular sludge internal circulation reactor. The International Water Association Anaerobic Digestion Model No. 1 extended with phosphorus (P), sulfur (S), and ethanol is used to describe the main biological and physico-chemical processes. The high-rate conditions within the reactor are simulated using a flow + reactor model comprised of a series of continuous stirred tank reactors followed by an ideal total suspended solids separation unit. Following parameter estimation by least squares on the measured data, the model had a relative mean error of 13 and 15% for data set #1 and data set #2, respectively. Response surfaces show that the reactor performance index (a metric combining energy recovery in the form of heat and electricity, as well as chemicals needed for pH control) could be improved by 45% when reactor pH is reduced down to 6.8. Model-based results reveal that influent S does not impose sufficient negative impacts on energy recovery (+5.7%, in MWh/day,+0.20 M€/year when influent S is removed) to warrant the cost of its removal (3.58 M€/year). In fact, the process could handle even higher S loads (ensuring the same degree of conversion) as long as the pH is maintained above 6.8. Nevertheless, a higher S load substantially increases the amount of added NaOH to maintain the desired operational pH (>25%) due to the acidic behavior of HS . CO 2 stripping decreases the buffer capacity of the system and hence use of chemicals for pH control. Finally, the paper discusses the possibilities and limitations of the proposed approach, and how the results of this study will be put into practice.

(Less)
Please use this url to cite or link to this publication:
author
organization
publishing date
type
Contribution to journal
publication status
published
subject
keywords
ADM1, benchmarking, energy recovery, high-rate reactors, response surfaces, sulfide production
in
Biotechnology and Bioengineering
pages
2726 - 2739
publisher
John Wiley & Sons
external identifiers
  • scopus:85053477515
ISSN
0006-3592
DOI
10.1002/bit.26807
language
English
LU publication?
yes
id
0ea83d74-09fb-4706-932c-7906dd5e5bc4
date added to LUP
2018-10-26 08:59:12
date last changed
2018-10-27 03:00:02
@article{0ea83d74-09fb-4706-932c-7906dd5e5bc4,
  abstract     = {<p>The objective of this paper is to present the model-based optimization results of an anaerobic granular sludge internal circulation reactor. The International Water Association Anaerobic Digestion Model No. 1 extended with phosphorus (P), sulfur (S), and ethanol is used to describe the main biological and physico-chemical processes. The high-rate conditions within the reactor are simulated using a flow + reactor model comprised of a series of continuous stirred tank reactors followed by an ideal total suspended solids separation unit. Following parameter estimation by least squares on the measured data, the model had a relative mean error of 13 and 15% for data set #1 and data set #2, respectively. Response surfaces show that the reactor performance index (a metric combining energy recovery in the form of heat and electricity, as well as chemicals needed for pH control) could be improved by 45% when reactor pH is reduced down to 6.8. Model-based results reveal that influent S does not impose sufficient negative impacts on energy recovery (+5.7%, in MWh/day,+0.20 M€/year when influent S is removed) to warrant the cost of its removal (3.58 M€/year). In fact, the process could handle even higher S loads (ensuring the same degree of conversion) as long as the pH is maintained above 6.8. Nevertheless, a higher S load substantially increases the amount of added NaOH to maintain the desired operational pH (&gt;25%) due to the acidic behavior of HS <sup>−</sup>. CO <sub>2</sub> stripping decreases the buffer capacity of the system and hence use of chemicals for pH control. Finally, the paper discusses the possibilities and limitations of the proposed approach, and how the results of this study will be put into practice.</p>},
  author       = {Feldman, Hannah and Flores-Alsina, Xavier and Kjellberg, Kasper and Jeppsson, Ulf and Batstone, Damien J. and Gernaey, Krist V.},
  issn         = {0006-3592},
  keyword      = {ADM1,benchmarking,energy recovery,high-rate reactors,response surfaces,sulfide production},
  language     = {eng},
  pages        = {2726--2739},
  publisher    = {John Wiley & Sons},
  series       = {Biotechnology and Bioengineering},
  title        = {Model-based analysis and optimization of a full-scale industrial high-rate anaerobic bioreactor},
  url          = {http://dx.doi.org/10.1002/bit.26807},
  year         = {2018},
}